There are large differences in the extent to which individuals abuse drugs, even when exposures are comparable. There are also differences in the effects drugs produce in individuals. It is not known, however, the extent to which an individual's propensity to abuse a drug, or self-administer it repeatedly, is related to the effects it produces. This study is designed to elucidate this relationship in order to increase our understanding of the mechanisms underlying individual differences in vulnerability to drug abuse. In the first part, subjects will participate in a 9-session self-administration experiment with 15 mg d-amphetamine and placebo, both administered orally. After sampling each of these two drugs during the first four sessions, subjects will be given a choice between them on five separate sessions. The degree of preference is measured by the number of times drug is chosen over placebo. In a second phase, changes in brain metabolism as assessed by PET and changes in brain electrophysiological activity in response to oral amphetamine in comparison to placebo will be assessed in order to correlate these effects with preference. Family history of drug use will also be determined and correlated with preference. To determine if any observed differences between those who tend to choose amphetamine and those who do not are due to the bioavailability of amphetamine, a third part will assess the pharmacokinetics of oral amphetamine in relationship to preference. Differences in hormonal responses will also be determined during this third part of the study. This study will help determine whether differences in propensity to self-administer a drug, which may indicate a vulnerability to drug abuse, are related to specific effects of the drug.